Numerous investigators have observed language disturbances among patients with schizophrenia. Language and speech abnormalities have been examined at 3 levels: semantic and syntactic structure, sound characteristics, and temporal patterning. The proposed study will examine the temporal aspects of speech in patients with schizophrenia, their biological relatives, and community subjects. Using an automated system developed by Cassotta et al. (1964) (AVTA), samples of semi-structured dialogue will be analyzed and the quality of vocal tracking assessed (the ability to track and adjust to the temporal characteristics of a partner?s speech). Vocal parameters produced by the AVTA system will be analyzed using time series regression (TRS). TSR produces a squared semi-partial cross-correlation that isolates and quantifies the extent to which a participant?s speech is influenced by or coordinated with that of another individual during a vocal interaction. This measure provides an objective and empirically validated assessment of "social give and take"--a uniquely disturbed aspect of the schizophrenia phenotype. (Measures of semantic/syntactic disturbance (Docherty et al., 1997) and sound characteristics (Alpert et al., 1995) will also be assessed. The project will examine the familial patterns of these language disturbances as well as their interrelations. These will be compared with known markers of genetic liability such as eye tracking and working memory deficits. The project will provide preliminary sensitivity and specificity estimates of identifying putative genetic risk using vocal tracking.)